4 ms·
Not exactly sure what you mean. After the model is trained there isn't much "machine learning abstraction". you can serve the model using tensorflow serving us
by quadrature 9y ago
Not exactly sure what you mean. After the model is trained there isn't much "machine learning abstraction".
you can serve the model using tensorflow serving using exported parameters. Its exactly what you want for serving the model.